Model becomes linear

Overfitting

Simpler interpretation

Underfitting

Polynomial regression can only be used for categorical data

Linear regression uses decision trees, while polynomial regression uses neural networks

Polynomial regression has fewer parameters than linear regression

Linear regression models relationships using straight lines, while polynomial regression models relationships using curves

Underfitting: The model is too simple to capture the true relationship.

High variance: The model fits training data well but generalizes poorly.

Overfitting: The model is too complex and captures noise in the data.

Data leakage: The model has seen information from the test set.

One column with the original categorical labels preserved

Three new binary columns, one for each color

One column containing integers 0, 1, or 2

Two columns, because one is redundant

The average price of a house in the dataset

The total price of a 1 m² house

The average number of square meters in the dataset

The increase in price for each additional square meter

To improve the model's accuracy on the training set

To evaluate how well the model performs on unseen data

To reduce the number of features in the model

To ensure the model is trained on all available data

y = 20 - 2*x

y = -20 + 2*x

y = -20 - 2*x

y = 20 + 2*x

    A       B       A      B

    A  B

----------------

    A  B

concat() got an unexpected keyword argument 'ignore_index'

There is no functional difference; pd.merge() and pd.concat() are interchangeable.

pd.merge() combines DataFrames along a particular axis without aligning on keys, while pd.concat() merges DataFrames based on common columns.

pd.merge() stacks DataFrames vertically, while pd.concat() stacks them horizontally.

pd.merge() combines DataFrames based on common column(s) or index(es), while pd.concat() joins DataFrames along a particular axis without considering column values.